ABSTRACT
The Optimal Power Flow problem (OPF) is an important problem of electrical engineering investigated since the 60s. The aim of the OPF problem is to determine theoperation point of a electricity transmission system that optimizes a given performance of such system and that satisfies its physical and operating constraints. The Reactive Optimal Power Flow problem (ROPF) is a particular case of the OPF problem. The ROPF problem can be mathematically modeled as a nonconvex nonlinear programming problem with discrete and continuous variables. In this paper, a new approach is proposed for solving theROPF problem. The proposed method comprises treating the discrete variables of the problem by a differentiable penalty function obtained by the decomposition of the triangular wave function through its Fourier series. The interior point methods implemented in the solver IPOPT is used to solve a generated sequence of continuous and penalized problems. The solution of the continuous and penalized problems converge to a solution of the original problem. Numerical tests with the IEEE 14 and 30 bus electrical systems are presented demonstrated the potential of the method.
Keywords:
optimal power flow; discrete variables; penalty function
